HIV infection leads to CD4 lymphopenia and immunosuppression, which can be successfully improved with antiretroviral therapy (ART) in the majority of people. Approximately 20 percent of HIV patients initiated on ART may develop an aberrant immune response known as immune reconstitution syndrome (IRIS). IRIS encompasses the paradoxical worsening of the manifestations of underlying opportunistic diseases (paradoxical IRIS) or the abrupt presentation of previously occult opportunistic disease (unmasking IRIS) in patients who recently started ART. Symptoms of IRIS can range from uncomplicated localized reactions to severe and systemic manifestations and, more rarely, autoimmune phenomena. Clinical studies have identified two main risk factors for IRIS: severe CD4 lymphopenia and the presence of opportunistic infections even if clinically silent (M. tuberculosis, M. avium complex, C. neoformans or other fungi, or viral pathogens) prior to ART initiation. The pathogenesis of IRIS in HIV infection remains unclear and there is no animal model that adequately mimics the clinical observations. In order to study the clinical predictors, biomarkers and pathogenesis of IRIS we have conducted a large prospective observational clinical trial of ART-naive HIV+ patients with severe CD4 lymphopenia (<100 cells/L) who were followed prospectively after initiation of ART. In collaboration with Dr Barber's laboratory in LPD we have also developed a murine model of mycobacterial IRIS that has highlighted the prominent role of Th1 responses in IRIS pathogenesis. In a previous study we evaluated the phenotype of T cells at baseline (pre-ART) and around the IRIS events. T cells from IRIS patients bore a highly activated phenotype (effector cells with high PD-1 expression) both pre-ART and at the time of IRIS, showing evidence of profound antigenic stimulation. IRIS patients also had a slower recovery of naive CD4 T cells following the IRIS events. These findings suggest a significant antigenic drive of T cell activation and expansion that becomes dysregulated as severe immunosuppression reverses. In continuation of that work we showed that the activated CD4 T cells predominantly recognize the underlying opportunistic pathogen and mount polyfunctional cytokine responses while the T cell responses to other pathogens, including HIV itself, remain intact. In a study of patients with HIV and chronic hepatitis co-infection with IRIS after ART initiation manifesting as a hepatitis flare, we showed evidence of a role of the antigenic load in IRIS pathogenesis as levels of hepatitis B viremia pre-ART and IL-10 levels were independent predictors of a hepatitis flare. Finally, in a different (autoimmune) form of IRIS that involves the thyroid and happens later after ART initiation, Grave's disease, we showed that recovery of naive CD4 T cells correlates with the appearance of anti-thyroid antibodies suggesting a very different pathogenesis than the infection-associated IRIS. More recently, we showed that monocytes play a prominent role in TB-IRIS and may serve as important predictors as well as potential therapeutic targets. Earlier findings from a pilot human study of HIV+ patients who were ART-naive and developed IRIS and comparing them with HIV+ patients who did not develop IRIS despite having similar opportunistic infections, suggest that two easily accessible tests from clinical laboratories (CRP and D-dimer) may be useful to predict or diagnose IRIS. This was further confirmed for CRP in a nested case-control study of a large clinical trial and has now been validated in a paradoxical tuberculosis (TB)-IRIS prospective study we performed with collaborators in Chennai, India. In a collaboration with the AIDS Clinical Trials Group (ACTG), we also found that levels of both Th1 (IFNgamma, TNF) and Th17 (IL-17) cytokines were elevated at baseline as well as IL-8, implying a possible role of neutrophils and myeloid cells that will now be further investigated. In a newer study we confirmed that monocyte and T cell markers such as sCD14 and gamma interferon, as well as D-dimer and vitamin D levels were independently associated with IRIS incidence in patients recruited in Mexico and South Africa. These studies will inform potential strategies for prevention or treatment of IRIS. Despite the significant improvement of morbidity and mortality in HIV infection in the ART era, mortality in HIV+ patients is still in excess of what is expected based on age and strongly relates to the degree of immunodeficieny as measured by CD4 T cell counts and to the degree of residual inflammation measured by IL-6, CRP, sCD14 and D-dimer. These markers appear to be linked more to activation of the innate system than the T cells. In addition, an activated monocyte phenotype was shown to be an independent predictor of progression of atherosclerosis in a cohort of well controlled treated HIV patients. We found that some markers of inflammation and coagulation are also elevated in HIV+ patients who control spontaneously the HIV viral load without taking anti-HIV medications. Finally, levels of IL-6 were independently associated with atherosclerosis and mortality in HIV infected persons with suppressed plasma viremia. Immune-based therapies and anti-inflammatory therapies are thus still being pursued with the objective to improve immune restoration and function in CD4 lymphopenic states. Newer strategies for decreasing residual inflammation in IRIS and in chronically treated patients are being pursued including commonly used medications such as statins and aspirin as well as phosphodiesterase 4 inhibitors.